How does a team with the best defensive record in the WSL lose a title on the final day? On Saturday at the Broadfield Stadium, Arsenal answered that question with ninety minutes of geometry that was too conservative for the occasion.

The arithmetic was simple before kick-off. Arsenal needed to beat Brighton to have any chance of overhauling Manchester City at the top of the table. A draw was not enough. A loss was catastrophic. Jonas Eidevall, whose side had conceded only 14 goals in 21 league matches, set up to control the game first and win it second. The contradiction between those two objectives defined the match.

The shape Arsenal chose

Eidevall named a 4-3-3 that, in possession, became a 2-3-5 in the attacking phase but only after a patient build-up that averaged 7.2 passes per sequence in the first half, according to the StatsBomb event data. The back four held a relatively deep line, averaging 39 metres from their own goal in the opening 30 minutes, which is 6 metres lower than their season average of 45 metres at the same phase of play. The midfield three of Frida Maanum, Lia Wälti, and Kim Little sat in a compact triangle, rarely stretching beyond 12 metres of vertical distance between the deepest and highest midfielder.

The intention was control. Arsenal had 68% of possession in the first half. They completed 287 passes to Brighton’s 112. They conceded zero shots in the opening 45 minutes. By every defensive metric, the plan was working. The problem was at the other end of the pitch.

The 23rd minute: the match in miniature

In the 23rd minute, the pattern that would define Arsenal’s afternoon appeared for the first time. Little received the ball from Wälti on the left side of the central circle. She had three options: a forward pass to Alessia Russo dropping into the space between Brighton’s midfield and defence, a switch to Beth Mead on the right wing, or a backward pass to Lotte Wubben-Moy in the centre-back pairing.

Little chose backward. The sequence continued: Wubben-Moy to Steph Catley at left-back, Catley back to goalkeeper Manuela Zinsberger, Zinsberger long to no one in particular. Brighton’s press had done nothing spectacular. They held a mid-block, average line at 42 metres, with the front two shadowing Arsenal’s double pivot. What they did was refuse to commit bodies forward. They did not chase the ball. They simply occupied the passing lanes that Arsenal needed to progress into the final third.

This happened 11 more times in the first half alone. Arsenal’s centre-backs completed 63 passes between them before the break. Only 7 of those were progressive (moving the ball at least 10 metres towards Brighton’s goal), per StatsBomb. The rest were lateral or backward. The geometry was circular. Arsenal kept the ball, and the ball kept Arsenal from scoring.

The half-space problem

Russo, Arsenal’s centre-forward, touched the ball 19 times in the first half. Of those, 14 were in the middle third of the pitch. She had 3 touches inside Brighton’s penalty area. Her average position was 11 metres deeper than her season heatmap median, which suggests she was coming short to find the ball because the service was not arriving in advanced areas.

On the right, Mead hugged the touchline but received only 6 passes in the first half. The half-space between Mead and Russo, the channel where Arsenal’s most dangerous combinations have lived this season, was occupied by Brighton’s left-sided centre-back, who stepped forward to intercept rather than dropping into the back line. Brighton’s centre-back pairing held a width of 28 metres, unusually wide for a back two, which compressed the half-spaces to roughly 9 metres each. Arsenal’s right-sided eight, Maanum, tried to find pockets in that 9-metre strip and succeeded twice, but neither instance led to a shot.

The goal that changed nothing

Arsenal scored on 51 minutes. Catley’s corner was flicked on at the near post and Russo stabbed home from 3 yards. The goal came from a set piece, not from open play, and that fact mattered. Brighton’s defensive block had not been broken. The geometry had not shifted. Arsenal had simply found a way to score without solving the structural problem.

The expected response was aggression. Push the defensive line higher. Stretch the pitch vertically. Force Brighton into one-on-one duels they could not win. Instead, Arsenal’s line dropped back. Their average defensive line height in the 15 minutes after the goal was 37 metres, lower than the first half. They protected the lead rather than extending it.

Brighton’s equaliser on 68 minutes was not a surprise. It was the logical consequence of a team that had been allowed to stay in the game. Brighton’s build-up had been patient and largely harmless, but Arsenal’s refusal to press the centre-backs in the second half gave Brighton time to find their wing-backs in advanced positions. The cross that led to the goal came from the right flank, where Brighton’s wing-back had 4.3 seconds on the ball before delivering. No Arsenal player closed the angle.

What Eidevall’s pragmatism cost

The defensive record is real. Arsenal conceded the fewest goals in the WSL this season, 15 in 22 matches. Their PPDA (passes allowed per defensive action) of 9.1 was the lowest in the league, meaning they pressed more aggressively per pass than any other side. But the pressing was front-loaded. It happened in the first half of the season and in the opening halves of matches. As the title race tightened, Eidevall’s Arsenal became a team that defended first and attacked as a consequence.

The numbers tell the story of drift. In their first 11 matches, Arsenal averaged 2.1 xG per game (StatsBomb model). In their last 11, that figure dropped to 1.3. Their progressive passes per 90 fell from 48 to 39 in the same split. The defensive structure held, but the attacking structure did not, and against a Brighton side that had nothing to play for except pride, the imbalance was terminal.

Manchester City, who clinched the title as Arsenal dropped points, scored 56 goals to Arsenal’s 48. They conceded 19 to Arsenal’s 15. The four-goal difference in attack outweighed the four-goal difference in defence, and it did so because City’s manager, Andrée Jeglertz, committed to a system that stretched opponents vertically even when protecting a lead. City’s average defensive line height in matches they led after 60 minutes was 44 metres. Arsenal’s was 38. Six metres of pitch is roughly one extra passing lane. Over a season, those lanes add up to goals.

The forward-looking constraint

Eidevall’s contract situation remains unresolved as of publication, per The Guardian. The tactical question for Arsenal, whoever is in charge next season, is structural. How do you maintain the best defensive record in the WSL while increasing attacking output by the 8 to 10 goals that separate second place from a title? The answer is not personnel. Russo, Mead, Maanum, and Caitlin Foord are among the most talented forwards in the league. The answer is geometry. The line has to go up. The half-spaces have to be stretched. The centre-backs have to be allowed to step, even when the lead is fragile.

On Saturday, Arsenal played not to lose. They got what they planned for.